# m33 ## Mouse Physiologic Parameters * Sex: Female * Strain: BALB/c * Weight (g): 20.7 * Vendor: Jackson Labs * Physical abnormalities: None * Protocol (https://cebs-ext.niehs.nih.gov/cahs/file/lapd/pages/Core/Animal%20Model.pdf) ## Aerosol Dosing * Particle size: 0.5 microns * Exposure time: 10 minutes * Exposure estimate: 9.90E+08 FMS | Time | RR (bpm) | Vt (ml) | VE (ml/min) | I:E | |:------------------|---------:|--------:|------------:|-----:| |Pre Aerosol | 233 | 0.24 | 56.3 | 0.85 | |Post 5min Aerosol | 222 | 0.21 | 46.0 | 0.77 | |Post 10min Aerosol | 195 | 0.19 | 37.4 | 0.64 | ## Slicing Info * Approximate lung orientation: LAS * Camera: D7100, 14-bit, 200mm Nikkor Macro lens, f/16 * Voxel size: 4.68 x 4.68 x 9.52 (microns) * Image exposures (msec). ISO=160, all images. | | mt | ol | fl | rd | Notes | |:--------:|:------:|:------:|:------:|:------:|:----------------------| |excitation| UVND2 | UVND2 | 485/20 | 560/20 |center wavelength/fwhm | |emission | N/A |470/30 | 535/30 | 635/30 | | |**Image** |**Exp** |**Exp** |**Exp** |**Exp** | **Notes** | |white | 40 | 250 | 8000 | 250 |Rhodamine B in OCT | |cal | 20 | 167 | 67 | 400 | | |images | | 77 | 20000 | 3000 | | |darks | | 77 | 20000 | 3000 | | * Notes regarding slicing: * Good aerosol signal. * Fl channel also looked okay. ## Airway Segmentation * Total centerline length: 506.133 mm * Number of branches: 1281 * Number of terminal branches: 641 * Maximum generation number: 26 * Number of outlet areas: 1677 ## Compartment Sizes and Aerosol Deposition * Lung volume: 1040.31 (mm^3) Compartments | Count | Volume (mm^3) ------------ | ----- | ------------- Lung | 1 | 1040.31 Lobes | 5 | 208.06±91.70 Sublobes | 57 | 17.92±18.70 Near acini | 327 | 3.05±1.86 Lobe | Volume (mm^3) | Average aerosol deposition --------- | ------------- | ---------------------- left | 351.08 | 1.11±1.90 cranial | 187.23 | 1.31±1.90 middle | 134.12 | 1.29±1.33 caudal | 269.16 | 1.09±1.44 accessory | 98.71 | 1.17±3.07 ## Additional Notes * Overall quality: B * Trachea was deformed and does not look physiologically correct; Segmentation does not model shape and does not attach to airway wall in some areas